Subpopulations have variable connections to specific institutions, such as the military, which can influence their use of social programs and access to resources. We use data from the 5-year (2008-2012) American Community Survey (ACS) public-use file to examine current Supplemental Nutrition Assistance Program (SNAP) use by military service status: active-duty personnel, recent veterans, long-term veterans, and reserve/guard members. Overall and by military service status, we estimate weighted descriptive statistics and multivariate logistic regression models that include demographic and socioeconomic controls. We document low but non-trivial levels of participation among active-duty personnel (2.2 %), higher but still moderate levels of SNAP use among veterans (7.1 % for recent veterans and 6.5 % for long-term veterans), and the highest level of use among members of the reserve/guard (9.0 %). Multivariate analyses support hypotheses based on the potential for the military, as a total institution, to substantially reduce use of SNAP among active-duty personnel, while veterans and reservists, who are more distal from food-related institutional resources, have higher likelihoods of using SNAP. Although levels of SNAP use among active-duty personnel, veterans, and reservists are lower than those observed in the national population, which includes those with no direct connection to military institutions, findings suggest that leaving active-duty military service results in a substantial and relatively immediate reduction in food-related resources for many recent veterans and their families. We discuss the implications of the findings for policy, limitations of the research, and directions for future research.